Species [ was: Re: Hadrosaur nomenclature]

["Jonathan R. Wagner" <jonathan.r.wagner@mail.utexas.edu>]

    Folks,

    Y'know, most systematists persist in discussing a "species problem." As
you will read, I think that "problem" is under control (perhaps not
resolved). To me, the most nagging, persistent, and pernicious issue is the
"genus problem."

    As a (student) paleontologist, I am frequently appalled at the ease with
which colleagues and outsiders blithely dismiss fossil species as
artifactual. This is usually done with the implication that the author is
being iconoclastic, or disabusing us neophyte paleontologists of our foolish
illusions, we who sit in our basements blindly and foolishly pantomiming the
(rigorous) systematic practices of "real" (neontological) systematists (hah!
idiots!). It has been pointed out time and again that, by whatever "species
concept" you use, neontological systematists do not often observe the
operable species criterion (e.g., breeding, speciation), and often "resort"
to the same methods paleontologists use (e.g., morphology). Granted, they
have more information, but the point of science is most certainly NOT to
simply abandon hope of progress in the face of missing data. So, from an
operational standpoint, such statements are not warranted.

    I think it is important to realize that the "biological species concept"
is not the only formulation of species, nor does it necessarily have a
corner on ontological reality. Indeed, De Queiroz (refs at bottom) has
suggested that all "species concepts" are little more than ways of
approaching a particular ontologically real entity, the lineage segment, and
that these "concepts" are not so much definitions of species as criteria for
their recognition. I find his arguments compelling, and I suggest you review
them before dismissing species as ontologically unreal.

    On that vein, recognizing species by any set of criteria will ALWAYS
"not work out," as HP Irmis states, in some cases. This is not a flaw in the
idea of species reality, but rather a flaw in our way of approaching the
world. The universe does not provide us with clear, distinct boundaries, and
reliance on legalistic adherence to verbal *definitions* will *never* allow
us to fully appreciate the variability of nature. Fuzzy boundaries are not
necessarily nonexistent boundaries, because sometimes real entities (like
the truth) point to themselves. As scientists, we all must come to terms, at
a very early age, with the (sometimes powerful) inconsistencies in the
concept of "individual organism," yet there are cases where most of us do
not hesitate to recognize these as real entities. If we change our focus
from the application of definitions to the recognition of diversity, we may
then recognize species, non-species, the "grey areas," where groups of
organisms are not "fully" species, and numerous other possible situations.

    As for the recognition of species in the fossil record, well, odds are
this will never be any more rigorous than it is in neontology... but just
because it is difficult (again) does not mean we should not attempt it.
Indeed, species reality suggests that working toward recognizing species in
the fossil record is a meaningful task. I actually have found some
discussions of variation in samples of dinosaurs, and their implications for
alpha taxonomy, to be very sophisticated. These might even approach the
truth at a level comparable to that achieved in many neontological studies.
Of course, there is no means of "proving" neontological species assignments
any more than there is for paleontological assignments. In the absence of
such a method, comparison, logic, parsimony, and, in general, science needs
to be applied in *both* situations. Certainly there will be more data in
neontological investigations. However, personally, I think that doesn't
always make it easier. I'd find it much easier suggesting that two nearly
morphologically identical dinosaurs found in close association at the same
horizon are the same species, than to have to deal with the grey wolf/ red
wolf situation, for example.

    As for conserving the binomial as a convenience, I must disagree
wholeheartedly. Our nomenclature should refer to real entities *only*. If I
believed that species or genera are not real, I feel it would be my duty to
abandon both (as some have suggested). Genera aren't real by anyone's
metric. We should chuck the idea along with all other categorical ranks.

    As for the "genus becoming the smallest recognizable unit of
biodiversity," I must say I run in *much* different circles than HP Yates.
Nearly every biologist I have ever heard/read/seen sees the species as this
unit. Indeed, given that many extant genera contain large numbers of species
(up to the thousands in beetles, I'm told), there does not seem to be a way
that this could be true. As a matter of fact, beyond association of a genus
with a particular clade, I have never heard, or heard of, *anyone* claiming
that genera in general are in any way "real." This is good, because they are
no more real than any other categorical rank of the Linnaean hierarchy.

    Now, in *paleontology*, The genus is a special case. It has been the
custom in paleontology to generate monotypic genera for new species. This is
a product of at least two, phylogenetic uncertainty and the mandatory nature
of the category. Often, there is the assumption that, because of the
dogmatically "incomplete nature" of the fossil record, monotypic genera are
appropriate because each fossil species "probably" represents a
"genus-level" assortment of unrecovered species. Even some sophisticated,
modern systematists readily accept this practice. With the added demand of
monophyly in higher taxa (and the genus still is one of these), some workers
appear to believe it is safer to name a new genus for each new species.
Naturally, to those who believe that genera should be "comparable," this is
anathema! However, the result is that there are a great number of monotypic
genera in paleontology; since these "genera" are operationally equivalent to
species, there is no problem in accepting them as units of biodiversity
(and, at least ideally, comparable to extant species).

    This convenient, if redundant, system is disrupted, however, by the
persistence of typological systematic methodology, wherein often small
morphological variations have been recognized as new species. An initially
monotypic genus often quickly becomes polytypic. Thus, a fossil genus often
circumscribes the level of variation expected of a modern species, or, to
put it more interpretively, represents a single species. However, the
vagaries of "conservative" taxonomy state that it must considered to be
several species. Thus it is then easier to just use the genus and ignore the
"species."

    It isn't so simple, however ... some genera contain a number of species
which are almost certainly one species, but a few species which are almost
certainly distinct (e.g., Chasmosaurus). Again, in recent years, the
approach has been to avoid the issue of species validity by using the genera
as a standard of comparison. Obviously, however, since genera are
biologically meaningless, this is an inappropriate comparison. Further, it
is operationally confusing. Statements are made, such as "the San Juan Basin
and Big Bend Early Maastrichtian ceratopsian faunas are different, because
they contain different genera, and the latter is closer to Alberta because
of the common occurrence of Chasmosaurus. Never mind that C. mariscalensis
in Big Bend is likely closer to P. sternbergii, and both are farther from C.
belli (which, incidentally, is the senior synonym of several other species).
Obviously, this is also an issue of generic monophyly. However, if C.
mariscalensis were given its own genus (as some would have done), the issue
would then become that the faunas are actually all very distinct, because
their each contain distinct genera of ceratopsians. Inclusion of
mariscalensis in Pentaceratops appears to be the solution, but then analysis
"at the generic level" fails to appreciate that it and sternbergii are, in
fact, different species.

    On a different tack, what exactly *is* comparable about generic
diversity anyone? By my count, there are seven species of hadrosaur in the
Judithian "Age" strata of Alberta, but, by the most restrictive count, only
six genera. Now, by species, I mean SPECIES, populations and groups of
populations connected lineally by ancestry and descent. Based on the
morphology of the crest, I'd say they weren't likely to have made little
hadrosaurs with other species. I wouldn't have, if I were them (and I spent
five years in Lubbock!). So then, how can the genus-level number possibly
compare with areas where the genus and species counts are identical? Can we
just ignore that one species, because it happens to cohabitate (to a very
modest degree) with a sister species? Should I just name it to a monotypic
genus, a simple paleosystematic prestidigitation that would, at a stroke of
the keys, remove it from obscurity to a meaningful role in the discussion of
biodiversity?

    As for the oft-heard cry that specific epithets are non-unique: WE HEAR
YOU. Everyone knows this, numerous proposals have been put forth to resolve
the problem. Personally, I note that if you use the WHOLE species name
(minus the generic epithet), that is, specific epithet, (species) author,
and year, nearly all such problems disappear. A simple page-order-based
consecutive lettering of the year will be sufficient to resolve the
remainder of the problem (e.g., _medius_ Linneaus 1759b). You can leave off
the latter parts of the name for convenience, and even add your favorite
clade-name to the front for clarity (e.g., _Iguanodontia mongoliensis_ vs.
_Ceratopsia mongoliensis_).


    To sum up:

    A) Species may be real.

    B) Whether or not they are, I disagree strongly with the contention that
fossil species are somehow different from neontological species, as they
involve recognizing the same class of real entities (although perhaps from
less data).

    C) Genera can be construed as the "minimum unit of biological diversity"
only in the context of current paleontological "lower" taxonomy, a construct
which itself is so confused that doing so only makes the problem worse.

    D) Genera are not only useless, but obfuscatory.

    E) Using genera as species is "sloppy" (if you'll pardon the severity of
the term), potentially obfuscatory, does not correspond to neontology, and
contributes nothing to systematics in paleontology or natural history in
general.

    F) The way to resolve issue of paleontological diversity is to address
the TRUE units of diversity, species. To do this, we must address named
fossil species from a critical standpoint, recognizing modern perspectives
on variation, and decide which species we consider likely to be real.

    G) We should relegate the "concept" of the genus to the wastebasket. The
names can (and should) be retained, but only to refer to clades, and only if
given unambiguous definitions.


Refs for those interested in the species issue (note, I think the title of
one of these *may* be wrong... ;)

De Queiroz, K. 1998. The general lineage concept of species, species
criteria, and the process of speciation, pp.57-75. In, D. J. Howard and S.
H. Berlocher (eds.), Endless Forms: Species and Speciation. Oxford
University Press, Oxford.

De Queiroz, K. 1999. The general lineage concept of species, species
criteria, and the process of speciation, pp.49-89. In, R. A. Wilson (ed.),
Species: New Interdisciplinary Essays. Massachusetts Institute of Technology
Press, Cambridge.


    That's my take on the issue, anyway.

    Wagner


Jonathan R. Wagner
9617 Great Hills Trail #1414
Austin, TX 78759